⚠️ AUTHORIZED USE ONLY This skill is for educational purposes or authorized security assessments only. You must have explicit, written permission from the system owner before using this tool. Misuse of this tool is illegal and strictly prohibited.
Mandatory confirmation gate Before running any command that probes, exploits, changes, persists on, extracts data from, or attempts credential access against a target:
- Ask the user to state the exact target URL, IP, account, or resource.
- Ask the user to confirm written authorization and the permitted scope.
- Show the exact command(s) and explain their expected effect.
- Wait for explicit confirmation in the current conversation.
Without that confirmation, remain read-only and provide defensive guidance only. Prefer a sandbox, disposable VM, or controlled lab.
When to Use
- An authorized assessment needs standard tooling orchestrated coherently.
- Choosing and sequencing the right tool per recon/exploitation phase.
Scope of Application
Use this skill when the task falls into the following scenarios:
- Target information gathering (port scanning, subdomain enumeration, service identification)
- Vulnerability scanning (Web vulnerabilities, CVE detection, misconfigurations)
- Web penetration (SQL injection, XSS, SSRF, directory brute-forcing)
- Password cracking (hash cracking, dictionary attacks)
- Network penetration (service exploitation, lateral movement assistance)
Division of Labor with Other Skills
| Scenario | What to Use |
|---|---|
| Active scanning/attacking (Nmap/Nuclei/SQLMap) | This skill |
| Reverse engineering binaries | ida-reverse/ or radare2/ |
| Frontend JS signature reverse engineering | js-reverse/ |
| Browser/desktop automation operations | browser-automation/ |
| CTF competitions (comprehensive) | CTF-Sandbox-Orchestrator/ |
Simple judgment:
- Need to "scan targets, discover vulnerabilities, exploit vulnerabilities" → this skill
- Need to "analyze internal program logic" → reverse engineering skills
- Need to "operate browser/desktop" → browser-automation
Tool Matrix
Information Gathering
| Tool | Purpose | Typical Command |
|---|---|---|
| Nmap | Port scanning, service identification, OS detection | nmap -sV -sC -O target |
| Masscan | Large-scale fast port scanning | masscan -p1-65535 target --rate=1000 |
| Subfinder | Subdomain enumeration | subfinder -d target.com |
| httpx | HTTP probing, liveness detection | httpx -l urls.txt -status-code |
Vulnerability Scanning
| Tool | Purpose | Typical Command |
|---|---|---|
| Nuclei | Template-based vulnerability scanning (CVE/configuration/exposure) | nuclei -u target -t cves/ |
| ZAP | Web application security scanning | Invoke via API or MCP |
| Nikto | Web server vulnerability scanning | nikto -h target |
Web Penetration
| Tool | Purpose | Typical Command |
|---|---|---|
| SQLMap | SQL injection automation | sqlmap -u "url?id=1" --batch --dbs |
| FFUF | Directory/parameter brute-forcing | ffuf -u target/FUZZ -w wordlist.txt |
| Gobuster | Directory/subdomain brute-forcing | gobuster dir -u target -w wordlist |
| XSStrike | XSS detection | xsstrike -u "url?param=test" |
Password Cracking
| Tool | Purpose | Typical Command |
|---|---|---|
| Hashcat | GPU hash cracking | hashcat -m 0 hash.txt wordlist.txt |
| John the Ripper | CPU hash cracking | john --wordlist=rockyou.txt hash.txt |
| Hydra | Online brute-force cracking | hydra -l admin -P pass.txt target ssh |
Exploitation Frameworks
| Tool | Purpose | Description |
|---|---|---|
| Metasploit | Vulnerability exploitation framework | Requires separate installation, large in size |
| Impacket | Windows protocol exploitation (SMB/WMI/Kerberos) | pip install impacket |
MCP Backend Selection
This skill supports two MCP backends; choose one:
Option A: pentestMCP (recommended, one-click Docker)
- Item：https://github.com/ramkansal/pentestmcp
- Features: 20+ tools packaged into a single Docker container, MCP server exposed directly
- Tools: Nmap, Nuclei, ZAP, SQLMap, FFUF, Nikto, Gobuster, Subfinder, httpx, etc.
- Installation：
# 拉取并运行
docker pull ramkansal/pentestmcp
docker run -d -p 8080:8080 ramkansal/pentestmcp
# 或本地构建
git clone https://github.com/ramkansal/pentestmcp.git
cd pentestmcp
docker build -t pentestmcp .
docker run -d -p 8080:8080 pentestmcp
- MCP Registration：
{
"mcpServers": {
"pentest": {
"url": "http://localhost:8080/mcp"
}
}
}
Option B: mcp-security-hub (modular)
- Item：https://github.com/FuzzingLabs/mcp-security-hub
- Features: Each tool is an independent MCP server, enabled on demand
- Tools：Nmap、Ghidra、Nuclei、SQLMap、Hashcat
- Installation: Follow the README of each submodule
Option C: Single-tool MCP (lightest)
If you only need one specific tool:
| Tool | MCP project | Installation |
|---|---|---|
| Nmap | nmap-mcp-server | npm |
| Nuclei | nuclei-mcp | npm |
| SQLMap | mcp-security-hub submodule | pip |
Reqable MCP (local packet capture and API workbench)
The Reqable desktop client can expose local packet capture, API, breakpoint, and rule capabilities through the official Reqable MCP Server . First install and start Reqable separately, then register the MCP:
powershell -NoProfile -ExecutionPolicy Bypass -File skills\scripts\bootstrap-reverse.ps1 -Capability reqable-mcp -McpHostTarget Codex
Replace Codex with Claude or Both to select the corresponding client; omitting -McpHostTarget will not write any client global configuration.
The registered stdio configuration is:
{
"mcpServers": {
"reqable-mcp": {
"command": "npx",
"args": ["-y", "reqable-mcp-server@1.0.1", "--scope", "minimal"]
}
}
}
- By default, Reqable's local API is used; if necessary, configure
--host、--portor--scope minimal|all。 minimalis the recommended default scope;allwill expose more tools that can change the proxy, rules, environment, or saved data.- For captured traffic, request replay, and rule modification, you must still first satisfy the authorization and network restrictions of
scope.md; the target scope must not be expanded just because the MCP has been registered.
Workflow
Standard penetration testing process
Important: When conducting penetration testing, you must run according to the autonomous loop framework of
references/pentest-loop.md. This framework defines the complete risk gating, logging standards, context compression, and completion check mechanisms.
1. 信息收集
- Nmap 端口扫描 → 确认开放服务
- Subfinder 子域名枚举 → 扩大攻击面
- httpx 存活检测 → 过滤有效目标
2. 漏洞扫描
- Nuclei 模板扫描 → 快速发现已知漏洞
- ZAP/Nikto → Web 应用深度扫描
3. 漏洞利用
- SQLMap → SQL 注入
- FFUF → 发现隐藏路径/参数
- 手动验证 → 确认可利用性
4. 后渗透（如果授权范围内）
- 权限提升
- 横向移动
- 数据提取
5. 报告
- 调用 docs-generator skill 生成渗透测试报告
Quick scan process (results in 5 minutes)
1. nmap -sV -sC target → 端口+服务
2. nuclei -u target -severity critical,high → 高危漏洞
3. 有 Web 服务 → ffuf -u target/FUZZ -w common.txt → 目录
4. 汇总发现 → 决定下一步
Notes
- Authorization is required — All scanning/attack operations must be within the authorized scope
- Control the scan rate — Avoid triggering WAF/IDS or crashing the target
- Passive first, then active — First information gathering, then vulnerability scanning, and finally exploitation
- Record all operations — Every command and result must be recorded for the report
- Do not automate blindly — AI should wait for confirmation at each key step
On-Demand Bootstrap
Automation capability boundaries
| Tool | Can be installed automatically | Installation method | Description |
|---|---|---|---|
| Nmap | ✓ | winget (Insecure.Nmap) |
Windows version |
| Nuclei | ✓ | go install or GitHub Release |
Requires Go or directly download the binary |
| SQLMap | ✓ | pip install sqlmap or git clone |
Python |
| FFUF | ✓ | GitHub Release | Go binary |
| SecLists | ✓ | GitHub Release ZIP | Dictionary collection (essential for FFUF/Gobuster) |
| Hashcat | ✗ | Manual download | Requires GPU driver |
| Metasploit | ✗ | Manual installation | Large in size; Kali is recommended |
| pentestMCP (Docker) | ✗ | Requires Docker | docker run ramkansal/pentestmcp |
| Impacket | ✓ | pip install impacket |
Python |
| ProxyCat | ✓ | pip install proxycat |
Proxy pool middleware (prevents blocking during batch scanning) |
| BurpSuite MCP | ✗ | Install from the BurpSuite extension marketplace | Requires BurpSuite Pro/Community |
| Reqable MCP | ✓ | npx -y reqable-mcp-server@1.0.1 |
The Reqable desktop client must be installed manually first |
Bootstrap strategy
- If the user has Docker → recommend pentestMCP (one-click all-in-one)
- If there is no Docker → install each tool separately as needed
- Prioritize installing Nmap + Nuclei + SQLMap (covering 80% of scenarios)
Manual installation guide
⚠️ **渗透工具未安装**
**推荐方案（需要 Docker）**：
docker pull ramkansal/pentestmcp
docker run -d -p 8080:8080 ramkansal/pentestmcp
**轻量方案（逐个安装）**：
- Nmap: winget install Insecure.Nmap
- Nuclei: go install -v github.com/projectdiscovery/nuclei/v3/cmd/nuclei@latest
- SQLMap: pip install sqlmap
- FFUF: 从 https://github.com/ffuf/ffuf/releases 下载
**安装后告诉我，我继续当前任务。**
Reference resources
- awesome-pentest — A comprehensive collection of 25k+ stars penetration testing tools
- SecLists — Dictionary/payload collections (essential for FFUF/Gobuster)
- PayloadsAllTheThings — Various vulnerability payloads
- HackTricks — Penetration technique encyclopedia
- pentest-ai-agents — 35 Claude Code pentest sub-agents (refer to their prompt patterns)
- Pentest Swarm AI — Swarm intelligence autonomous pentest framework (multi-agent collaboration, supports MCP server)
- ProxyCat — Proxy pool middleware (prevent IP blocking during batch scanning)
- planning-with-files — Scheduled task skill (for loop testing)
Reference documents within this skill
references/pentest-loop.md— Core loop framework(risk gating + logging standards + context compression)references/recon-pipeline.md— Authorized reconnaissance pipeline(CF headers / nmap / Evidence)references/client-side-lab-playbook.md— DOM XSS / prototype pollution / agent-browser(range client side)references/burpsuite-mcp-guide.md— BurpSuite MCP complete guide(63 tools + 7 major usage scenarios + AI Prompt templates)references/automation-loop-pattern.md— Automated loop testing mode (lightweight version)references/awesome-pentest-digest.md— Pentest tool essentials quick referencereferences/pentest-ai-agents-matrix.md— 35 agent coverage matrixpayloads/— Custom payload directory (AI uses preferentially)templates/— Required pentest file templates (scope/rules/plan/findings/progress)
src-hunter vulnerability discovery knowledge base
src-hunter/ The directory contains a complete SRC/Bug Bounty vulnerability discovery methodology:
- 19 categories of attack playbooks(IDOR, RCE, XSS, SQLi, SSRF, OAuth, file upload, etc.)
- 305 structured payloads + 263 WAF/EDR bypass steps
- 2887 disclosed HackerOne High/Critical reports
- 88,636 WooYun historical case statistics
- Domestic component fingerprints and default credentials
- CVSS 4.0 report template
Usage: AI automatically reads the corresponding playbook during the hunt phase and tests according to its workflow.
See src-hunter/SKILL.md and src-hunter/references/。
Routing context
Upstream entry: skills/SKILL.md(master control),routing.md
Trigger conditions: Requires actively scanning/attacking the target (port scanning, vulnerability detection, injection testing, etc.)
Downstream exit:
- Web vulnerability found requiring further analysis →
js-reverse/ - Binary vulnerability found requiring reverse engineering →
ida-reverse/orradare2/ - Browser operation required to verify vulnerability →
browser-automation/ - Generate report after completion →
docs-generator/
Peer associated modules: CTF-Sandbox-Orchestrator/(Web/Pwn challenges in CTF will use these tools)
Task completion self-check (MUST pass before claiming completion)
- Did I execute every step in the workflow (rather than just reading)?
- Did I use a real tool path based on
tool-index? - Did I produce reproducible evidence (commands/scripts/screenshots/reports)?
- Did I complete and write back the Checklist items required by RULES?
Limitations
- Tool availability varies by platform; verify installs before engagements.
- Noisy scanning can disrupt fragile targets; tune timing and scope.
Adapted from zhaoxuya520/reverse-skill (MIT).